Feature Store Summit 2022
Check out all videos and slides presented at the conference!
Accelerating ML at Uber with the Palette Feature Store
Amit Nene from Uber shows how the Michelangelo ML Platform uses the Palette Feature Store to addresses inefficiencies of the model lifecycle.
Get the presentationHopsworks Feature Store after 4 years: Lessons learned and what's next
Moritz Meister and Fabio Buso from Hopsworks share the lessons learned over the past four years of building their platform and what lies ahead.
Get the presentationBuilding A Feature Store For Hyper Growth
Brian Seo goes over Doordash's feature store architecture, learnings from supporting CRDB and what it takes to build a feature store.
Get the presentationFeathr: An Enterprise-Grade, High Performance Feature Store
David Stein from Linkedin and Xiaoyong Zhu from Microsoft cover the background of Feathr, its core concepts and design, and their journey on scaling an enterprise FS.
Get the presentationPanel Discussion - APIs for Feature Stores
Moderated by Jim Dowling from Hopsworks. A discussion on the historical evolution and the future of APIs for feature stores.
Get the presentationNexus Feature Store powering Disney Magic
Dustin Hamerla from Disney Streaming describes his team's journey towards building Nexus, an in-house feature store and how it accelerates feature engineering.
Get the presentationIntroducing Zipline: An Open Source Feature Engineering Platform
Nikhil Simha from Airbnb introduces Zipline, a declarative feature engineering platform developed at Airbnb, which will be open-sourced in March.
Get the presentationFighting against Marketplace Scams using Community-driven AI
Sinan Ozdemir covers how Shiba uses a Feature Store to maintain ML models with up-to-date data on how bad actors target communities.
Get the presentationFeature Store Observability: What it is, and why it matters
Aman Khan from Arize discusses the state of ML production monitoring, its challenges, and how to actively improve models and features in production.
Get the presentationPanel Discussion - Data Quality & Feature Stores
Moderated by Patrik Liu Tran from Validio. A discussion on challenges of data quality for features and whether differentiate from general data quality requirements for analytics.
Get the presentationHamilton: an open source, declarative, micro-framework for clean & robust feature transform code in Python
Stefan Krawczyk from Stitch Fix presents Hamilton an open source Python micro framework that solved his team's pain points by changing their working paradigm.
Get the presentationAmerica First Credit Union optimizes their MLOps stack with the Hopsworks Feature Store
Richard Woolston elaborates on how AFCU's adoption of the Hopsworks Feature Store has helped them significantly improve their workflows.
Get the presentationFast Sub-ML use-case development using feature stores
Achnit Thomas from Scribble Data talks about Sub-ML, a class of applications simpler than traditional ML approaches and often used in decision support systems.
Get the presentationFeature Store Usage Patterns - From a Single Data Scientist to an Enterprise
Simba Khadder from Featureform shares the team's learnings from different companies on usage patterns of feature stores.
Get the presentationPanel Discussion - How to make the Feature Store become an invisible part of the productionalization of ML models
Moderated by Sarah Catanzaro from Amplify Partners. A discussion on the challenges of making the feature store disappear and become part of the workflow of data science and data engineering.
Get the presentationPowering Enterprise Feature Stores with a Universal Semantic Layer
Gaurav Rao from AtScale talks about how enterprises can apply the power of the semantic layer to enrich feature stores and scale business-ready AI.
Get the presentationEmpowering your feature stores with AI feature discovery
Lulu Liu from dotData will discuss how Feature Discovery and Feature Factory concepts can transform your feature development process.
Get the presentationScaling Feature Engineering with Dagger
Ravi Suhag from Open DataOps Foundation talks about Dagger and hot it can be used with feature stores to empower data scientists to make feature engineering self-service.
Get the presentationDesigning Feature DSLs: Principles and Tradeoffs
Greg Kuhlmann from Sumatra discusses feature designs and describe their journey developing a DSL for streaming feature transformation.
Get the presentationralf: Real-time, Accuracy Aware Feature Store Maintenance
Sarah Wooders, PhD UC Berkeley introduces the notion of feature store regret that helps evaluate feature quality of different maintenance policies.
Get the presentationOpenMLDB: An Open-Source Real-Time Feature Platform Computing Consistent Features for Training and Inference
Lu Mian from 4Paradigm introduces OpenMLDB, an open-source ML database that provides a real-time feature platform for ML applications that reduces dev cost.
Get the presentation